The broad goals of this project are to improve self-care (e.g., medication use) and health outcomes among older adults with inadequate health literacy. Elders with chronic illness such as hypertension have less effective health literacy skills compared to younger adults, which predicts poor self-care and outcomes. While the association of health literacy and health is often presumed to reflect poor comprehension of health information, there is little direct evidence for these relationships, in part because health literacy models are not integrated with language comprehension theory. Health literacy in the present project is conceptualized in terms of a cognitive model of language processing that identifies the multiple levels of representing information conveyed by text. Specific aims are: (1) To model components of health literacy by examining the extent to which individual differences in standard measures of health literacy are explained by general measures of cognitive ability (i.e., working memory) and health-related knowledge. (2) Based on this model, to investigate relationships between components underlying health literacy and processes involved in understanding medication and discharge instructions related to hypertension. This involves a) Exploring the text demands involved in constructing multileveled representations of these texts; b) Assessing how individual differences in health literacy (and its cognitive underpinnings) impact different levels of these representations, so as to contribute to variation in comprehension, memory, and problem-solving from health texts; (3) To investigate the impact of critical text characteristics (e.g., pictorials) on literacy differences in performance. The first stage of the research will use a broad battery of measures of fluid mental abilities and health knowledge to model health literacy as a combination of patient abilities critical to comprehension. Studies in the second stage build on this ability-based model of health literacy to investigate differences in comprehension and memory for health texts. Regression analyses will investigate how these differences in performance reflect cognitive ability and health knowledge. The final studies investigate how to design texts to mitigate the comprehension difficulties associated with cognitive and knowledge limits. The project will produce guidelines for constructing health texts for older adults with different profiles of cognitive resources and knowledge, mitigating effects of inadequate health literacy on health outcomes. Public Health Relevance: Findings from this project will integrate health literacy models with theories of language comprehension and cognitive aging to help explain the impact of health literacy on patients' ability to understand, remember, and use health information. This knowledge in turn will guide design of health care texts such as medication and hospital discharge instructions, so that these texts better guide self-care behaviors, thereby mitigating the impact of inadequate health literacy on health behaviors and outcomes.